两类品种工件混流的多站点CSPS系统优化控制
发布时间：2019-08-03 20:25
【摘要】：研究一种两类品种工件混流的多站点传送带给料加工站系统的优化控制问题.系统中的站点如何协同工作完成工件加工任务,是提高系统生产率的重要课题.将前视距离作为各站点的决策变量,通过站点间的局部信息交互,提出一种品种均衡工作模式,并运用一种模型无关的串行反馈式多agent强化学习算法求解系统的最优策略.实验结果验证了该工作模式的合理性和算法的有效性,并分析了部分参数变化对系统性能的影响.
[Abstract]:In this paper, the optimal control problem of a multi-station transmission feed processing station system with mixed flow of two kinds of workpiece is studied. How to work together to complete the workpiece processing task is an important task to improve the productivity of the system. Taking the forward looking distance as the decision variable of each site, a variety equilibrium working mode is proposed through the local information interaction between stations, and a model independent serial feedback multi agent reinforcement learning algorithm is used to solve the optimal strategy of the system. The experimental results verify the rationality of the working mode and the effectiveness of the algorithm, and analyze the influence of some parameter changes on the performance of the system.
【作者单位】： 合肥工业大学电气与自动化工程学院;
【基金】：国家自然科学基金面上项目(61174186,61573126,71231004) 教育部高等学校博士学科点专项科研基金项目(20130111110007);教育部新世纪优秀人才计划项目(NCET-11-0626) 合肥工业大学应用科技成果培育计划项目(JZ2016YYPY0052)
【分类号】：TP278
本文编号：2522779
[Abstract]:In this paper, the optimal control problem of a multi-station transmission feed processing station system with mixed flow of two kinds of workpiece is studied. How to work together to complete the workpiece processing task is an important task to improve the productivity of the system. Taking the forward looking distance as the decision variable of each site, a variety equilibrium working mode is proposed through the local information interaction between stations, and a model independent serial feedback multi agent reinforcement learning algorithm is used to solve the optimal strategy of the system. The experimental results verify the rationality of the working mode and the effectiveness of the algorithm, and analyze the influence of some parameter changes on the performance of the system.
【作者单位】： 合肥工业大学电气与自动化工程学院;
【基金】：国家自然科学基金面上项目(61174186,61573126,71231004) 教育部高等学校博士学科点专项科研基金项目(20130111110007);教育部新世纪优秀人才计划项目(NCET-11-0626) 合肥工业大学应用科技成果培育计划项目(JZ2016YYPY0052)
【分类号】：TP278
【相似文献】
相关期刊论文 前1条
1 唐昊;裴荣;周雷;谭琦;;基于状态聚类的多站点CSPS系统的协同控制方法[J];自动化学报;2014年05期
相关硕士学位论文 前4条
1 盛文学;基于工件位置信息的多站点CSPS系统的协同控制[D];合肥工业大学;2015年
2 李博川;两品种工件混流的多站点CSPS系统最优控制研究[D];合肥工业大学;2016年
3 裴荣;基于状态聚类和需求驱动的多站点CSPS系统的协同控制[D];合肥工业大学;2013年
4 王超;多工序CSPS系统基于柔性站点的分层优化控制研究[D];合肥工业大学;2013年
，本文编号：2522779
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2522779.html
